Market sizing · M04 · Sunflower Intelligence

Before you start

You will be able to

Define a market size by its products, customers, place, period and unit, and convert volume into value at a stated price level.
Distinguish the total addressable, serviceable available and serviceable obtainable markets (TAM, SAM and SOM), stating the definitions you use.
Calculate a market or sales estimate from the top down (category × shares) and from the bottom up (outlets × rate of sale, or customers × buying × frequency), in Excel.
Apply an explicit conversion assumption to stated purchase intent, and explain why intent is not behaviour.
Evaluate an estimate by triangulation, sensitivity analysis and sanity checks, and work out what would have to be true for a goal to be reached.

Review first

The product life cycle — M02 set Lantern Cold Brew’s first-year goal of about 100,000 cans in a category of 4.8 million; this lesson tests it.
Situation analysis — Market share and substitutes, which define what a market includes, were introduced there.
Survey error — The February survey’s weighted intent figures and the caution about stated intentions come from that lesson.
What-if analysis: data tables, Goal Seek, scenarios and Solver — The sizing model uses Goal Seek, a data table and sensitivity analysis.

Fictional case A fictional coffee business founded by Ana Lim. It starts as a catering and barista-training service, opens its own cafés and an online shop for beans, and plans a canned cold brew.

The idea

Fictional case

It is March 2027. Lantern Cold Brew is still a plan: 250 ml cans at $3.49 in Harbour City, where about 4.8 million cans of ready-to-drink coffee are sold a year. M02’s plan set a first-year goal of about 100,000 cans. Since then Ana has talked to two store buyers, asked her mailing list about Cold Brew in the February survey and audited competitors’ shelf prices. Before she orders a first production run, her board wants to know how many cans Lantern can really expect to sell, and whether 100,000 is an estimate or a hope.

What a market size is

A market size is the quantity of a product that a defined group of customers buys in a defined place over a defined period. All four parts matter. “The cold brew market” means nothing until you say which products count (canned ready-to-drink coffee only, or bottles and café iced coffee too), whose purchases count, where and over how long. Harbour City’s 4.8 million cans is a yearly figure for canned ready-to-drink coffee, all brands and all channels; change any part of the definition and the number changes, and so does every market share calculated from it.

Size can be measured in volume (units: cans, tubs, memberships, seats) or in value (money), and value is volume times an average price. Say which price. Shoppers pay the shelf price; producers receive the shelf price less what the shops and distributors keep. At $3.49 on a supermarket shelf, Lantern receives about $2.37 a can (M02).

TAM, SAM and SOM

Market sizes are often given as three nested figures. The total addressable market (TAM) is the whole demand for the product category in the place and period, whoever supplies it. The serviceable available market (SAM) is the part of the TAM that the business’s offer and its routes to market could serve. The serviceable obtainable market (SOM) is the part of the SAM that the business can realistically win in the period, given its resources and its competitors.

The labels are most used in start-up and technology practice, and sources define them differently. TechTarget describes TAM as the revenue opportunity at a 100% market share, “as if no competition exists”; Carta describes it as the total demand for your product across all potential customers. For one source SAM is the part of the TAM a company’s products can serve, for another the part it can reasonably expect to reach, and some put a time limit on SOM while others don’t. So say in words what each of your three figures counts. This course uses the definitions in the previous paragraph.

Lantern Cold Brew’s market as a funnel (not to scale). The TAM is M02’s 4.8 million cans; the SAM is the 70% sold in the three kinds of outlet in Lantern’s plan; the SOM is Ana’s bottom-up estimate, beside M02’s goal. Top-down estimates work down the funnel; bottom-up estimates build up from stores and customers.

For Lantern Cold Brew the TAM is M02’s 4.8 million cans a year. The SAM is the cans sold through the three kinds of outlet in Lantern’s plan: supermarkets (45% of the category’s cans), natural and specialty grocers (15%) and cafés and online shops (10%), the kind of outlet Lantern runs itself. Together they sell 70% of the category, 3.36 million cans; the convenience stores’ 30% is outside it because the plan leaves them out. The SOM is the figure the board is asking for.

Two routes to an estimate

Top-down market sizing starts from the whole category and narrows it with shares: the cans sold in the planned channels, then the share of those the business might win. Bottom-up market sizing builds the figure from its parts: how many stores will stock the product and how many it sells in each, or how many customers will buy and how often. Top-down is quick and shows the scale. Bottom-up is slower, but every input is something a store buyer, a customer or a trial can confirm or contradict.

Ana’s bottom-up estimate for year one is 48,200 cans: 30 supermarkets at 10 cans a store a week, 25 natural grocers at 14, and 1,200 cans a month through Lantern’s own cafés and online shop. M02’s goal is 100,000. One is 1.4% of the SAM and the other 3.0%. This lesson is about deciding which to believe, and what would have to be true for the larger one.

In depth

Drawing the boundary

The first decision is which products count. Draw the line too tightly and you miss real competition; draw it too loosely and the market looks huge while your share looks tiny. M01’s substitutes give the test: a product belongs in the market if the customers you want treat it as a real alternative.

Four ways to draw Lantern Cold Brew’s market (only the first is measured in the case)
Market definition	What it adds	Useful for
Canned ready-to-drink coffee, Harbour City, a year (4.8 million cans)	The category Lantern enters	Shares, shelf space, the launch plan
Ready-to-drink coffee in any pack	Bottles and cartons	A brand that might add bottles, or shoppers who switch between packs
Cold coffee bought away from home	Iced coffee from cafés and kiosks	Lantern’s own cafés, and long-term growth
Cold caffeinated drinks on the go	Energy drinks and iced tea	Rarely: too broad to guide a launch

The broader definitions aren’t wrong; they answer different questions. A launch plan needs the category the product will be shelved in, because that is where shoppers compare it and where store buyers judge it.

TAM, SAM and SOM in practice

A TAM can be measured two ways. Where the category already sells, count what it sells now, as Lantern’s 4.8 million cans do. Where buyers have no current product to count, or you want the ceiling, count every potential customer at the offer’s price (customers × quantity × price), TechTarget’s “100% market share”; it is usually far larger, so say which you used.

The TAM is narrowed to the SAM by whatever limits what the business could serve: the channels it will sell through, the regions it can deliver to, the customer types its offer suits. Lantern’s SAM is limited by channel. A stricter analyst would count only the stores expected to list Lantern, which gives a much smaller SAM (an exercise asks you to work it out). Neither choice is wrong, but the shares calculated from them differ, so say which you used.

The SOM has to come from evidence about what the business can win: listings, rates of sale, customers who will buy. It is an estimate. A goal is different: a target the business sets itself. When a goal is larger than the SOM, the plan has to change, or the goal does.

Lantern Cold Brew, year one: the three figures and the goal (fictional case)
Figure	What it counts	Cans	Share of TAM
TAM	Canned ready-to-drink coffee, all brands and channels	4,800,000	100%
SAM	Cans sold in supermarkets, natural grocers and cafés and online shops	3,360,000	70%
SOM	Ana’s bottom-up estimate of Lantern’s first year	48,200	1.0%
Goal	M02’s first-year target (not an estimate)	100,000	2.1%

Top-down: from the category to a share

A top-down estimate multiplies the category by a chain of shares. Its inputs usually come from secondary data: trade associations, syndicated data from retail panels, industry reports and official statistics (the secondary-research lesson covers where to find them and how to judge them; OpenStax gives an industry report of all related product sales as an example). Lantern’s chain is short: 4.8 million × 70% = 3.36 million cans in the SAM. The last link, the share of the SAM that Lantern will win, is the hard one, and nothing in the category data says what it should be.

Bottom-up: from outlets or from customers

For a product sold in shops, a bottom-up estimate multiplies the number of outlets that will stock it by its rate of sale, the units an outlet sells in a period, usually per store per week. Outlet counts come from store lists, distributors and official business statistics: in the United States, the Census Bureau’s County Business Patterns gives the number of establishments with paid employees by six-digit industry code for every county, each year. Rates of sale come from store buyers, from comparable products and, best of all, from a trial in a few stores.

For a product sold direct, or to customers the business can count, build from customers instead: the number of people in the target group, the share who will buy, how often they buy and how much each time. The same structure fits a service (members × uptake × visits × price) and a business product (firms × adoption × seats × price per seat).

Ana’s bottom-up estimate for Lantern Cold Brew’s first year (fictional case)
Route	Stores listing	Cans a store a week	Periods	Cans a year
Supermarkets	30 of 120	10	52 weeks	15,600
Natural and specialty grocers	25 of 45	14	52 weeks	18,200
Lantern’s cafés and online shop		1,200 cans a month	12 months	14,400
Total				48,200

The listings and rates of sale are Ana’s planning assumptions after talking to two store buyers; the direct figure is her estimate of what Lantern’s own cafés and online shop will sell, informed by what subscribers said in the February survey, not a calculation from it. Each is a number that someone can challenge, which is the point of building from the bottom.

Volume into value

To turn cans into dollars, multiply by an average price, and choose it with care. The right average is weighted by volume, the price of the average can actually sold, which needs each brand’s sales. Ana’s shelf-price audit (the next lesson) gives 24 shelf prices from ten stores, averaging $3.15, so the category is worth about 4.8 million × $3.15 ≈ $15.1 million a year at shelf prices. The average of the three brands’ own averages is $3.05, which gives $14.6 million. If the cheapest brand, Daily Grind, sells the most cans (it is the best known, at 75% awareness in M02), the true figure is lower still.

Value at shelf prices is what shoppers spend; producers receive less. In M02’s three retail channels the distributors and retailers keep 32% to 40% of the shelf price. Weighting those margins by each channel’s share of the category, and assuming the cafés and online shops keep nothing (true only of a brand selling through its own, as Lantern will), the trade keeps 32.1%, so producers receive at most about $10.3 million. Quote the figure that matches the decision, and say which it is.

Using what customers say they will do

Surveys give another bottom-up route: ask people how likely they are to buy, then count buyers. In February Ana asked her mailing list, on a five-point purchase-intent scale, how likely they would be to buy Lantern Cold Brew at about $3.49 in the next three months. Weighted to the list’s four groups by post-stratification (the survey-error lesson explains the weights), 13.7% said they definitely would and 24.6% probably would: about 819 and 1,475 of the 6,000 people on the list.

Stated intent isn’t behaviour: the survey-error lesson showed that it tends to overstate buying, most of all for new products. Armstrong, Morwitz and Kumar (2000) found that forecasts built on purchase intentions for existing products were more accurate than extrapolating past sales, but they note that intentions-based methods often give biased estimates, overstating or understating actual buying, and that sales data should be used to correct the bias where they exist. A new product has no sales of its own, so the correction is an assumption: a conversion rate for each answer, borrowed where possible from a comparable launch, and varied to see what it does.

The list is roughly the people Lantern’s own cafés and online shop already reach (café customers who never joined it are missing), so it suits the direct channel, provided those who buy do so there rather than in a supermarket, or they would be counted twice. Suppose 40% of those who said “definitely” and 10% of those who said “probably” become regular buyers, each taking a four-pack a month (fictional assumptions). That gives 819 × 0.40 + 1,475 × 0.10 ≈ 475 regular buyers and about 1,900 cans a month, against Ana’s 1,200.

Now vary the assumption. At 25% and 5% the estimate is about 1,114 cans a month; at 60% and 20%, about 3,146. The conversion rates move the answer by a factor of nearly three.

The intent route puts the own channel at about 1,900 cans a month against Ana’s 1,200, so her figure is the more cautious; even at 3,146 a month the bottom-up total would be about 71,500 cans, still short of the goal. Notice also what the question measured: intent to try within three months. How many of those who try come back every month is a separate assumption that no intent question answers.

Triangulating: what a gap means

Triangulation in sizing means estimating the same quantity by independent routes and comparing the results (the secondary-research lesson does the same with sources). Agreement is reassuring only if the routes really are independent. Disagreement is more useful, because it points at the assumption that differs. Lantern has two estimates for year one, and a target.

Two estimates and a target for Lantern Cold Brew’s first year
Figure	Cans	What it assumes
Ana’s bottom-up estimate	48,200	55 listings at the buyers’ rates of sale; 1,200 cans a month direct
M02’s first-year goal	100,000	A target, set before any store was asked
M02’s Marketing Mix Simulator, default plan	199,000	A teaching model in which every store in a chosen channel stocks Lantern

The simulator seems to support the goal until you look at what it assumes. It treats a channel as all or nothing: once Lantern chooses supermarkets, it is on every supermarket’s shelf. Spread over all 120 supermarkets, its 92,139 supermarket cans come to about 14.8 cans a store a week, and its natural-grocer cans to about 37 a store a week, against the buyers’ 10 and 14.

Put Ana’s 55 listings into the simulator’s rates of sale and its figure falls to about 91,000 cans; use her direct estimate as well and it is about 86,000. The rest of the gap is the rates of sale. How many stores will list Lantern, and how fast it sells in each, are the two things to research next.

The gap matters for money as well as volume. M02’s pricing lesson put the default plan’s break-even at about 117,000 cans a year. On Ana’s estimate, the first year wouldn’t cover its fixed costs, and even the goal is below break-even; whether that is acceptable for a launch year is the board’s judgement, but the sizing makes it visible.

Sensitivity, ranges and sanity checks

A sizing model has many inputs, and they are not equally important. A sensitivity analysis moves one input at a time from a low to a high value you can defend, holding the others at their central values, and ranks the inputs by how much they move the answer. Ana’s low and high values, her own judgements after the buyer conversations (fictional), give the table below.

Sensitivity of Ana’s bottom-up estimate (central value 48,200 cans), largest swing first
Assumption (central value)	Low–high	Cans at low	Cans at high	Swing
Natural grocers: cans a store a week (14)	8–20	40,400	56,000	15,600
Own channel: cans a month (1,200)	600–1,800	41,000	55,400	14,400
Supermarkets: cans a store a week (10)	6–14	41,960	54,440	12,480
Supermarkets listing Lantern (30)	20–40	43,000	53,400	10,400
Natural grocers listing Lantern (25)	18–30	43,104	51,840	8,736

The rate of sale in natural grocers matters most: natural grocers supply the largest part of Ana’s estimate (18,200 cans), and she is unsure of their rate (8 to 20 cans). Her range for the own channel is wider still in relative terms, but that channel supplies less. The ranking depends on the ranges, so they must be honest, not chosen to make a point.

With every input at its low value the model gives 20,928 cans, and with every one at its high value 81,920. Those extremes are unlikely, but not because errors cancel: the three rates of sale all depend on how shoppers take to the can, so they may well miss together. With the listings as planned and every rate at its low value the model gives 26,960 cans; at its high value, 69,440. Even that is below 100,000.

Report the result as a range around a central figure and name what drives it: “about 48,000 cans, 40,000 to 56,000 if any one assumption is off, and most sensitive to how fast it sells in natural grocers.” Then sanity-check the figure against what you already know.

The category. An average Harbour City supermarket sells about 346 ready-to-drink coffee cans a week (2.16 million ÷ 120 ÷ 52), so Ana’s 10 cans would be 2.9% of a listing store’s category, plausible for a new brand. In natural grocers 14 of about 308 is 4.55%.
Capacity. Can the co-packer make the volume, and can a small team serve every store that lists the product?
Units and periods. Weeks against months, cans against four-packs, a calendar year against a launch year that starts part-way through it.
Double counting. A subscriber who buys in the online shop and in a supermarket appears in two routes.

A figure that fails a sanity check is wrong somewhere, even if every multiplication is right.

Formulas

Market value from volume

V = Q \times \bar{p}

$V$: Market value over the period$
$Q$: Market volume over the periodunits
$\bar{p}$: Average price per unit sold, at a stated level: the shelf price, or what producers receive$ per unit

Assumes. The average is weighted by volume (the price of the average unit sold). A simple average of shelf prices is only a stand-in.

Producers’ value = shelf value × (1 − the trade’s share of the shelf price).

Top-down sizing (a chain of shares)

Q_{\text{SAM}} = a \times Q_{\text{TAM}}, \qquad Q_{\text{SOM}} = m \times Q_{\text{SAM}}

$Q_{\text{TAM}}$: Total addressable market: the category’s sales, all suppliers and channelsunits or $
$a$: Share of the TAM the business’s offer and routes to market could serve
$Q_{\text{SAM}}$: Serviceable available marketunits or $
$m$: Share of the SAM the business expects to win in the period
$Q_{\text{SOM}}$: Serviceable obtainable market: the estimate for the businessunits or $

Assumes. Each share applies to the figure before it, for the same products, place and period.

With several filters (channel, region, customer type), a is their product. The share m needs evidence of its own.

Bottom-up sizing from outlets

Q = W \sum_{c} L_c \, v_c

$Q$: Units sold in the periodunits
$W$: Weeks in the periodweeks
$L_c$: Outlets in channel c that stock the productoutlets
$v_c$: Rate of sale in channel c: units an outlet sells a weekunits per outlet per week
$c$: A channel

Assumes. Every listing outlet in a channel sells at the channel’s average rate, every week of the period.

Add sales through any direct channel separately (cans a month × 12 for Lantern).

Bottom-up sizing from customers

Q = N \times r \times f \times u

$Q$: Units sold in the periodunits
$N$: Customers in the target grouppeople or firms
$r$: Share of them who buy in the period
$f$: Purchases per buyer in the periodpurchases
$u$: Units per purchaseunits

Assumes. The share who buy and their frequency apply to the whole target group; segments that differ are summed separately (SUMPRODUCT).

Multiply by price for value; for a service, u is often 1 and f the number of visits.

Buyers from stated intent

B = N \sum_{k} c_k \, p_k

$B$: Expected buyerspeople
$N$: People the survey representspeople
$p_k$: Share giving answer k (weighted to the population)
$c_k$: Conversion assumption: share of those giving answer k who will actually buy
$k$: An answer on the intent scale (definitely, probably …)

Assumes. The c_k are assumptions, ideally from comparable launches or past sales; they are not measured by the survey.

Show the result at low and high values of the c_k.

Rate of sale needed to reach a goal

v^{*} = \dfrac{G - Q_{\text{other}}}{L \times W}

$v^{*}$: Rate of sale needed in one channelunits per outlet per week
$G$: The goal for the periodunits
$Q_{\text{other}}$: Units expected from every other routeunits
$L$: Outlets listing the product in that channeloutlets
$W$: Weeks in the periodweeks

Assumes. The other routes deliver what the plan assumes.

Goal Seek finds the same value in a model. Solve for L instead to find the listings needed.

Worked examples

A first example: Sizing a plant-based ice cream from the top down

Fictional figures

Quarrymoor Creamery (fictional) plans an oat-milk ice cream in 500 ml tubs for its home region. A trade report it bought says that shoppers in the region buy 9.0 million tubs of ice cream a year, 6% of them plant-based. The creamery will sell only through supermarkets and farm shops, which sell 80% of the plant-based tubs. Plant-based tubs average $6.50 on the shelf, and the shops keep 35% of the shelf price. Size the market in tubs and in dollars, and work out what 5% of the serviceable market would be worth to the creamery in a year.

Define the market first. The creamery competes with other plant-based ice creams, so take the category as plant-based tubs in the region over a year. TAM: $9{,}000{,}000 \times 0.06 = 540{,}000$ tubs. Defining it as all ice cream would give 9.0 million tubs and make every share look tiny.
Narrow to the channels it will use (the top-down chain): SAM $= 540{,}000 \times 0.80 = 432{,}000$ tubs.
Convert to value at a stated price ( $V = Q \times \bar{p}$ ). At shelf prices: $432{,}000 \times 6.50 = 2{,}808{,}000$ dollars. At the creamery’s net price, after the shops’ 35%: $6.50 \times 0.65 = 4.225$ dollars a tub, so $2{,}808{,}000 \times 0.65 = 1{,}825{,}200$ dollars.
Apply the share: 5% of the SAM is $432{,}000 \times 0.05 = 21{,}600$ tubs, worth $21{,}600 \times 4.225 = 91{,}260$ dollars a year to the creamery.

Answer. TAM 540,000 tubs; SAM 432,000 tubs, worth $2.81 million at shelf prices or $1.83 million to producers; 5% of the SAM is 21,600 tubs, about $91,000 a year to Quarrymoor.

What it means. The top-down chain gives the scale in four multiplications, but its last input, 5%, came from nowhere. Before the creamery plans production it should build the same figure from the bottom up: how many supermarkets and farm shops will stock the tub, and how many each sells a week.

A harder example: Three estimates for a gluten-free loaf

Fictional figures

Tillworth Bakery (fictional) has six shops in its town and is deciding whether to bake a gluten-free loaf every day. It has three ways to estimate a year’s sales. (1) The town has 21,000 households; a regional shopper survey (fictional) found that 7% of households buy gluten-free bread, 1.2 loaves a week on average, and Tillworth sells about 15% of the bread the town buys. (2) In a two-week trial in its two busiest shops the loaf sold 40 and 34 a week, and those shops serve 1.4 times as many customers as the average Tillworth shop. (3) Its loyalty scheme has 2,600 members and records 55% of the bakery’s sales; 9% of members bought the trial loaf, about once a fortnight. Estimate the year’s loaves by each route, and decide what to plan for.

Top-down (category × shares): $21{,}000 \times 0.07 = 1{,}470$ households buy gluten-free bread; $1{,}470 \times 1.2 \times 52 = 91{,}728$ loaves a year in the town, the TAM. At Tillworth’s usual 15%: $91{,}728 \times 0.15 \approx 13{,}759$ loaves.
Bottom-up by shop (outlets × rate of sale × weeks): the trial shops averaged $(40 + 34) \div 2 = 37$ loaves a week. An average shop would sell about $37 \div 1.4 \approx 26.4$ , so six shops for 52 weeks give $26.43 \times 6 \times 52 \approx 8{,}246$ loaves.
Bottom-up by customer (customers × share who buy × frequency × units): $2{,}600 \times 0.09 = 234$ members buying, 26 times a year, one loaf each: $6{,}084$ loaves. Members account for 55% of sales, so if other customers behave alike the total is $6{,}084 \div 0.55 \approx 11{,}062$ loaves.
Read the gap. The two bottom-up routes give about 8,200 and 11,100; the top-down gives 13,800. The top-down assumes Tillworth wins gluten-free buyers at its usual share of all bread, but many such households may buy from supermarkets’ free-from shelves, which nothing here measured. The shop route may still be high, because a two-week trial catches curious first-time buyers; the customer route assumes that customers outside the scheme buy like its members. The customer route rests on the same two-week trial, so it isn’t independent of the shop route, and it assumes every member who tried the loaf buys it every fortnight all year, a repeat rate the trial couldn’t measure.
Decide. Plan on the bottom-up range, about 8,000 to 11,000 loaves a year (roughly 160 to 210 a week across the six shops), and treat 13,800 as an upper bound until the share has been tested.

Answer. Top-down about 13,759; by shop about 8,246; by customer about 11,062 loaves a year. Plan for about 8,000 to 11,000 (160 to 210 a week), not 13,800.

What it means. The routes don’t agree, and that is informative: the gap sits on one assumption, the share of gluten-free buyers who would buy from a bakery. A month of real sales, compared week by week, will test it better than more desk research, so the bakery can start at the lower volume and revise after the first month.

Applied: a business decision: What would have to be true for 100,000 cans?

Fictional figures

M02’s plan set a first-year goal of about 100,000 cans of Lantern Cold Brew. Ana’s bottom-up estimate, after talking to two store buyers, is 48,200: 30 of Harbour City’s 120 supermarkets listing Lantern at 10 cans a store a week, 25 of its 45 natural and specialty grocers at 14, and 1,200 cans a month through Lantern’s own channel. Her board asks whether the goal is realistic and, if not, what would have to be true to reach it.

Place both figures in the funnel (share of SAM = Q ÷ SAM). Goal: $100{,}000 \div 3{,}360{,}000 = 3.0\%$ . Bottom-up: $48{,}200 \div 3{,}360{,}000 = 1.4\%$ . The gap, $100{,}000 - 48{,}200 = 51{,}800$ cans, is larger than the whole bottom-up estimate.
Rates of sale (the required-rate formula). Holding the listings and the own channel, the stores must sell $100{,}000 - 14{,}400 = 85{,}600$ cans a year, $85{,}600 \div 52 \approx 1{,}646$ a week across the 55 stores, against $300 + 350 = 650$ in Ana’s plan: $2.53$ times her rates of sale in every store. If only the supermarkets sped up, they would need $(100{,}000 - 18{,}200 - 14{,}400) \div (30 \times 52) \approx 43.2$ cans a store a week.
Listings. At 10 cans a week a supermarket adds $10 \times 52 = 520$ cans a year, so with the natural grocers as planned the goal needs $67{,}400 \div 520 \approx 129.6$ supermarkets: more than the 120 in Harbour City. With all 45 natural grocers listing ( $45 \times 14 \times 52 = 32{,}760$ cans), it still needs 102 of the 120.
Own channel. On its own it would need $(100{,}000 - 15{,}600 - 18{,}200) \div 12 \approx 5{,}517$ cans a month, nearly a can a month for every one of the 6,000 people on Lantern’s mailing list.
Sanity-check the most plausible route. 43.2 cans a week would be 12.5% of an average supermarket’s 346 category cans, for a brand known to about 12% of category buyers. The sensitivity table agrees: with every assumption at its high value the model reaches only 81,920 cans.

Answer. On the evidence, 100,000 cans is a hope, not an estimate. It needs Lantern to sell 2.5 times as fast as the buyers expect in every listing store, or to be listed in more supermarkets than Harbour City has, or to sell nearly a can a month to everyone on its list. The bottom-up estimate is about 48,000 cans (1.4% of the SAM).

What it means. Asking what would have to be true turned an argument about optimism into three conditions that can be checked. The board can now put the right questions to store buyers (how many stores, how fast) and fund a trial of rates of sale in a few stores. It also shows what is at stake: with M02’s break-even at about 117,000 cans, the first year is an investment on either figure.

Explore

Common mistakes

“Our TAM is the opportunity.”
The TAM counts every sale in the category, by every competitor, through every channel. A business can win only part of its SAM, and only with listings, customers and time. Lead with the SOM and show how it was built; quote the TAM for context.
“One per cent of a big market is a safe assumption.”
A share chosen to sound modest is still an assumption, and for a new brand 1% can be hard. One per cent of Harbour City’s category is 48,000 cans, which is Ana’s whole plan: 55 listings at the buyers’ rates of sale plus 1,200 cans a month direct. Build the share from listings and rates of sale, or from customers, rather than picking it.
“The goal is our forecast.”
A goal is a target; an estimate is what the evidence supports. When the two differ, as 100,000 and 48,200 do, work out what would have to be true for the goal, check it, and change either the plan or the goal.
“Customers who say they will buy are buyers.”
Stated intent isn’t behaviour, and intentions-based forecasts can overstate or understate buying. Apply an explicit conversion rate to each answer, vary it, and remember that a question about the next three months measures trial, not repeat purchase.
“A single number looks more professional.”
A point estimate hides how uncertain it is. Give a central figure, a range and the assumption that moves it most: that tells the reader what to watch and what to research next.
“Volume times the average shelf price is the producers’ revenue.”
Shelf prices measure what shoppers spend. Producers receive the price after the trade’s margin, and a simple average of shelf prices isn’t weighted by what sells. State the price level and how the average was formed.

Guided practice

Work through these with hints when you need them. Every hint and every worked solution is there to be used.

Loading four questions…

On your own

No hints this time. Answer, check, and read the feedback; the solution appears once you have answered or if you ask for it. Your first answers count toward mastery.

Loading six questions…

In practice

Three people use market sizes in three different ways. An investor reading a start-up’s pitch looks past the TAM, which is usually large and easy to find, to how the SAM is defined and to the bottom-up build of the SOM: how many customers, at what price, reached through which channel. A pitch that offers only “1% of a huge market” hasn’t done the work.

A retailer’s category buyer, deciding whether to list a new product, thinks in rates of sale: how many units a store a week, against the products already on the shelf, and what the space and the listing fee would earn otherwise. A supplier that can show credible rates of sale from a trial, rather than a share of the national market, gets a more serious hearing.

A product manager uses the range, not the central figure, to plan production and stock. A first run sized to the central estimate, with a contract that allows a quick second run, keeps both too little and too much stock affordable. The assumption with the widest swing tells her what to measure first, usually in a small test market or a trial in a few stores.

Fictional case

Ana’s note to her board, March 2027: “Our bottom-up estimate for Cold Brew’s first year is about 48,000 cans, 1.4% of the cans sold in our planned channels, and 40,000 to 56,000 if any one assumption is off. The 100,000-can goal would need us to sell 2.5 times as fast as the store buyers expect in every store that lists us, or to be listed in more supermarkets than Harbour City has. I propose we replace the goal with this range, size the first production run to the central figure, and measure rates of sale in a few stores before ordering a second. Whether to launch on these numbers is for our April review.”

The same discipline applies to a bakery planning a new loaf, a software firm pricing an add-on or a gym chain opening a site: define the market, build the estimate from parts someone can check, compare routes, and report a range with the assumption that drives it.

Knowledge check

Six questions, one attempt each, no hints. Score 80% or more to complete the lesson; you can take it again at any time.

Loading six questions…

Summary

A market size is the quantity of a defined product bought by defined customers in a defined place and period, in units (volume) or money (value); change the definition and the size, and every share, change.
TAM is the whole category, SAM the part the business’s offer and channels could serve, SOM the part it can realistically win; sources define them differently, so say what each of yours counts.
Top-down multiplies the category by shares, and its weak link is the share you assume you will win. Bottom-up builds from outlets × rate of sale × weeks, or customers × share who buy × frequency × units.
Value = volume × average price: say which price (shelf or producers’), and weight the average by volume where you can.
Stated intent needs an explicit conversion assumption for each answer; vary it, and don’t mistake intent to try for repeat buying.
Triangulate: estimate by independent routes and read a gap as a pointer to the assumption that differs.
Run a sensitivity analysis, report a central figure with a range and its main driver, and sanity-check against the category, capacity, units and double counting.
Ask what would have to be true for a goal: Lantern’s 100,000 cans needs 2.5 times the buyers’ rates of sale, or more supermarkets than Harbour City has.

Terms introduced

Market size: The quantity of a product bought by a defined group of customers in a defined place over a defined period, measured in units (volume) or money (value).
Total addressable market: TAM: the whole demand for a product category in a defined place and period, whoever supplies it, measured either as what the category sells now or, as a ceiling, as every potential customer buying at the offer’s price. Definitions vary between sources; some describe it as the revenue available at a 100% market share.
Serviceable available market: SAM: the part of the total addressable market that a business’s offer and routes to market could serve, such as the sales in the channels it plans to use. Definitions vary between sources.
Serviceable obtainable market: SOM: the part of the serviceable available market that a business can realistically win in a stated period, given its resources and competitors. An estimate, not a goal.
Top-down market sizing: Estimating a market, or a business’s sales, by starting from the whole category and narrowing it with shares: the channels or segments served, then the share the business expects to win.
Bottom-up market sizing: Estimating a market, or a business’s sales, by building it from its parts: outlets × rate of sale × periods, or customers × the share who buy × how often × how much.
Rate of sale: The number of units of a product an outlet sells in a period, usually per store per week; also called velocity.

Sources

TechTarget (2023). “What is TAM SAM SOM?” Written in 2019 and revised by TechTarget editors in 2023; definitions of TAM, SAM and SOM and of top-down and bottom-up sizing. techtarget.com/enterprise-software/definition/TAM-SAM-SOM
The Carta Team (2023, 14 April). “Market size: TAM, SAM, & SOM”. Carta. carta.com/blog/market-size
Armstrong, J. S., Morwitz, V. G. & Kumar, V. (2000). Sales forecasts for existing consumer products and services: do purchase intentions contribute to accuracy? International Journal of Forecasting, 16(3), 383–397. The sentences on bias are in the authors’ working paper (MPRA paper 81689, April 2000). mpra.ub.uni-muenchen.de/81689/1/MPRA_paper_81689.pdf
U.S. Census Bureau. “County Business Patterns” (programme overview: establishments, employment and payroll by six-digit NAICS industry, for the nation, states, counties, metropolitan areas and ZIP codes, annually since 1964). census.gov/econ/overview/mu0800.html
U.S. Census Bureau. “About County Business Patterns (CBP)” (annual statistics for establishments with paid employees). census.gov/programs-surveys/cbp/about.html
Gomez Albrecht, M., Green, M. & Hoffman, L. (2023). Principles of Marketing, section 6.3, “Steps in a Successful Marketing Research Plan” (internal and external secondary data, including an industry report of all related product sales). OpenStax. openstax.org/books/principles-marketing/pages/6-3-steps-in-a…

Before you start

You will be able to

Review first

The idea

What a market size is

TAM, SAM and SOM

Two routes to an estimate

In depth

Drawing the boundary

TAM, SAM and SOM in practice

Top-down: from the category to a share

Bottom-up: from outlets or from customers

Volume into value

Using what customers say they will do

Triangulating: what a gap means

Sensitivity, ranges and sanity checks

Formulas

Market value from volume

Top-down sizing (a chain of shares)

Bottom-up sizing from outlets

Bottom-up sizing from customers

Buyers from stated intent

Rate of sale needed to reach a goal

Worked examples

Explore

Common mistakes

“Our TAM is the opportunity.”

“One per cent of a big market is a safe assumption.”

“The goal is our forecast.”

“Customers who say they will buy are buyers.”

“A single number looks more professional.”

“Volume times the average shelf price is the producers’ revenue.”

Guided practice

On your own

In practice

Knowledge check

Summary

Key formulas

Terms introduced

Sources